To understand the biophysical basis of the diffusion MR signal, Uri Nevo, a former STBB post-doctoral fellow, and now an Assistant Professor at Tel Aviv University, has successfully constructed and tested an experimental system to interrogate organotypic cultured brain slices using diffusion MRI methods. This work has already shown promising results relating changes in the measured diffusion coefficient map to changes in environmental conditions to which the cultured tissue is subjected. One hypothesis that emerged from these studies is that active processes occurring at many different length scales (cell streaming, water flow across membranes, etc.) are responsible for some signal loss in the diffusion weighted MRI signal. This insight prompted the tandem development of a theory to explain how microscopic fluid flows affect the measured diffusion weighted MRI signal and the diffusion coefficient measured in tissues (i.e., pseudo-diffusion) and an experimental model system, the Rheo-NMR, in which well-characterized flow fields can be produced that create known amounts of pseudo-diffusion. The importance of these combined studies is that if such microscopic motions, like streaming, water flow across membranes, etc., manifest themselves as additional signal loss in diffusion weighted MRI experiments, then we can use information about this additional signal loss to infer different aspects of cell function and vitality, including excitability. We have also been collaborating with Bradley Roth to try to examine different physical mechanisms that could be exploited or used to detect neural currents directly using MRI. One approach we examined previously was whether small displacements caused by Lorentz forces produced in strong magnetic fields (like those within a large clinical MRI scanner) could be employed to measure neural currents in vivo using MRI. Our calculations showed that the induced displacements of nerves caused by Lorentz forces would not be detectable by MRI using existing technology. In the area of Transcranial Magnetic Stimulation (TMS), Pedro Miranda and his group in Lisbon, in association with STBB, has performed detailed calculations using finite element methods (FEM), to predict the electric field and current density distributions induced in the brain during TMS. Previously, we found that both tissue heterogeneity and anisotropy of the electrical conductivity (i.e., the conductivity tensor field) contribute significantly to distort the induced fields, and even to create excitatory or inhibitory hot spots in some regions. More recently, we have been developing FEM models of cortical folds, containing gyri and sulci. We showed that this complicated cortical anatomy also significantly affects the distribution of induced electric fields within the tissue, and the location and types of nerve cells that could be excited or depressed by such stimuli. These phenomena could have significant clinical consequences both in interpreting or inferring the region or locus of excitation and in determining the source of nerve excitation. We are beginning to marry our macroscopic models of TMS with microscopic models of nerve excitability in the CNS to be able to predict the locus of excitation in TMS and even the populations of neurons that are excited. Recently, we have also been applying these advance FEM models to study the possible physical mechanism underlying Direct Current Excitation (DCE).